Model fusion aims to integrate the capabilities of multi-expert models into a single model. Existing evaluations offer limited coverage of reasoning with long chains of thought and tool use, while strong average performance can mask substantial losses in individual expert capabilities. We introduce FusionEval, a benchm...
Shuo Cai, Yang-Gan Gu, Yuhang Liu et al.· 0 citations
Native NVFP4 promises to accelerate reinforcement learning (RL) for large language models (LLMs), but achieving stable learning across separate training and sampling engines remains challenging. We show that matching model weights and numerical formats is insufficient to ensure consistent policy evaluation, highlightin...
Zhen Li, Ming-Fa Feng, Yi-Ming Zhang et al.· 0 citations
This paper characterize the interaction between mismatch and the policy-gradient direction, distinguishing locally amplifying from contracting update contributions that mismatch magnitude alone cannot identify and introduces TRIAGE, a direction-aware stabilization method that uses segment-level diagnosis to selectively...
Zhen Li, Shuai Zhang, Yang-Gan Gu et al.· 1 citation
SMAT (Simple MAT) is introduced, which jointly optimizes expert loss and expected loss at simulated merged parameters generated by sampling scaling coefficients, masks, and additive noise, and improves the mean score across five merging methods.
Yang-Gan Gu, Yuan-Yi Wang, Zhen Li et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.